#!/usr/bin/env ruby
require 'rails-html-sanitizer'

class I18nExtractor
  ERB_TAG = /<%(.*?)%>/
  HTML_TAG = /<[^>]*>|<\/[^>]*>|[^>]*\/>|\".+">|">|<\w+(.+|)|.+="(.+)|"/
  SEPARATOR = '_@@@_'
  SKIP_TAGS = [[/<script/i,/<\/script>/i],[/<%/,/%>/],[/<style/i,/\/style>/i]]

  def initialize(filename)
    @filename = filename
    @stack = []
    @files_with_results = []
  end

  def extract
    sanitizer = Rails::Html::FullSanitizer.new
    i = 0
    File.foreach(@filename) do |line|
      i += 1
      next if in_script_block?(line)
      striped_line = sanitizer.sanitize(line).chomp.strip.gsub(HTML_TAG, '')
      next if striped_line.empty?
      if striped_line.length > 1
        unless @files_with_results.include?(@filename)
          @files_with_results << @filename
          printf "\n\n\e[4m#{@filename}\e[00m\n"
        end
        printf("%3s:%s\n",i+1,striped_line)
      end
    end
  end

  private

  def in_script_block?(s)
    return true if s.nil? || s.strip.size == 0 
    jump_in_tag = SKIP_TAGS.find{ |start_tag,end_tag| s =~ start_tag}
    @stack.push jump_in_tag[1] if jump_in_tag
    if @stack.last
      end_tag_match = s.match(@stack.last) 
      if end_tag_match
        @stack.pop 
        return in_script_block?(end_tag_match.post_match)
      end
    end
    return !@stack.empty?
  end

end

printf "\nSeeking out erb files that need translating. Just one moment...\n\n\e[1mThis script definitely includes false negatives and false positives, but hopefully it helps!!!\e[00m\n\n\n"
Dir.glob("**/*.html.erb").each do |filename|
  next if filename.include?("/admin/") # We are not yet focused on this area
  I18nExtractor.new(filename).extract
end